Journal articles: 'Deep learning architecture'

1

Munir, Khushboo, Fabrizio Frezza, and Antonello Rizzi. "Deep Learning Hybrid Techniques for Brain Tumor Segmentation." Sensors 22, no. 21 (October 26, 2022): 8201. http://dx.doi.org/10.3390/s22218201.

Full text

Abstract:

Medical images play an important role in medical diagnosis and treatment. Oncologists analyze images to determine the different characteristics of deadly diseases, plan the therapy, and observe the evolution of the disease. The objective of this paper is to propose a method for the detection of brain tumors. Brain tumors are identified from Magnetic Resonance (MR) images by performing suitable segmentation procedures. The latest technical literature concerning radiographic images of the brain shows that deep learning methods can be implemented to extract specific features of brain tumors, aiding clinical diagnosis. For this reason, most data scientists and AI researchers work on Machine Learning methods for designing automatic screening procedures. Indeed, an automated method would result in quicker segmentation findings, providing a robust output with respect to possible differences in data sources, mostly due to different procedures in data recording and storing, resulting in a more consistent identification of brain tumors. To improve the performance of the segmentation procedure, new architectures are proposed and tested in this paper. We propose deep neural networks for the detection of brain tumors, trained on the MRI scans of patients’ brains. The proposed architectures are based on convolutional neural networks and inception modules for brain tumor segmentation. A comparison of these proposed architectures with the baseline reference ones shows very interesting results. MI-Unet showed a performance increase in comparison to baseline Unet architecture by 7.5% in dice score, 23.91% insensitivity, and 7.09% in specificity. Depth-wise separable MI-Unet showed a performance increase by 10.83% in dice score, 2.97% in sensitivity, and 12.72% in specificity as compared to the baseline Unet architecture. Hybrid Unet architecture achieved performance improvement of 9.71% in dice score, 3.56% in sensitivity, and 12.6% in specificity. Whereas the depth-wise separable hybrid Unet architecture outperformed the baseline architecture by 15.45% in dice score, 20.56% in sensitivity, and 12.22% in specificity.

APA, Harvard, Vancouver, ISO, and other styles

2

Alvarez-Gonzalez, Ruben, and Andres Mendez-Vazquez. "Deep Learning Architecture Reduction for fMRI Data." Brain Sciences 12, no. 2 (February 8, 2022): 235. http://dx.doi.org/10.3390/brainsci12020235.

Full text

Abstract:

In recent years, deep learning models have demonstrated an inherently better ability to tackle non-linear classification tasks, due to advances in deep learning architectures. However, much remains to be achieved, especially in designing deep convolutional neural network (CNN) configurations. The number of hyper-parameters that need to be optimized to achieve accuracy in classification problems increases with every layer used, and the selection of kernels in each CNN layer has an impact on the overall CNN performance in the training stage, as well as in the classification process. When a popular classifier fails to perform acceptably in practical applications, it may be due to deficiencies in the algorithm and data processing. Thus, understanding the feature extraction process provides insights to help optimize pre-trained architectures, better generalize the models, and obtain the context of each layer’s features. In this work, we aim to improve feature extraction through the use of a texture amortization map (TAM). An algorithm was developed to obtain characteristics from the filters amortizing the filter’s effect depending on the texture of the neighboring pixels. From the initial algorithm, a novel geometric classification score (GCS) was developed, in order to obtain a measure that indicates the effect of one class on another in a classification problem, in terms of the complexity of the learnability in every layer of the deep learning architecture. For this, we assume that all the data transformations in the inner layers still belong to a Euclidean space. In this scenario, we can evaluate which layers provide the best transformations in a CNN, allowing us to reduce the weights of the deep learning architecture using the geometric hypothesis.

APA, Harvard, Vancouver, ISO, and other styles

3

Kumar, Bhavesh Shri, Naren J, Vithya G, and Prahathish K. "A Novel Architecture based on Deep Learning for Scene Image Recognition." International Journal of Psychosocial Rehabilitation 23, no. 1 (February 20, 2019): 400–404. http://dx.doi.org/10.37200/ijpr/v23i1/pr190251.

Full text

APA, Harvard, Vancouver, ISO, and other styles

4

Hyunhee Park, Hyunhee Park. "Edge Based Lightweight Authentication Architecture Using Deep Learning for Vehicular Networks." 網際網路技術學刊 23, no. 1 (January 2022): 195–202. http://dx.doi.org/10.53106/160792642022012301020.

Full text

Abstract:

<p>When vehicles are connected to the Internet through vehicle-to-everything (V2X) systems, they are exposed to diverse attacks and threats through the network connections. Vehicle-hacking attacks in the road can significantly affect driver safety. However, it is difficult to detect hacking attacks because vehicles not only have high mobility and unreliable link conditions, but they also use broadcast-based wireless communication. To this end, V2X systems need a simple but a powerful authentication procedure on the road. Therefore, this paper proposes an edge based lightweight authentication architecture using a deep learning algorithm for road safety applications in vehicle networks. The proposed lightweight authentication architecture enables vehicles that are physically separated to form a vehicular cloud in which vehicle-to-vehicle communications can be secured. In addition, an edge-based cloud data center performs deep learning algorithms to detect car hacking attempts, and then delivers the detection results to a vehicular cloud. Extensive simulations demonstrate that the proposed authentication architecture significantly enhanced the security level. The proposed authentication architecture has 94.51 to 99.8% F1-score results depending on the number of vehicles in the intrusion detection system using control area network traffic.</p> <p> </p>

APA, Harvard, Vancouver, ISO, and other styles

5

Hao, Xing, Guigang Zhang, and Shang Ma. "Deep Learning." International Journal of Semantic Computing 10, no. 03 (September 2016): 417–39. http://dx.doi.org/10.1142/s1793351x16500045.

Full text

Abstract:

Deep learning is a branch of machine learning that tries to model high-level abstractions of data using multiple layers of neurons consisting of complex structures or non-liner transformations. With the increase of the amount of data and the power of computation, neural networks with more complex structures have attracted widespread attention and been applied to various fields. This paper provides an overview of deep learning in neural networks including popular architecture models and training algorithms.

APA, Harvard, Vancouver, ISO, and other styles

6

Zou, Han, Jing Ge, Ruichao Liu, and Lin He. "Feature Recognition of Regional Architecture Forms Based on Machine Learning: A Case Study of Architecture Heritage in Hubei Province, China." Sustainability 15, no. 4 (February 14, 2023): 3504. http://dx.doi.org/10.3390/su15043504.

Full text

Abstract:

Architecture form has been one of the hot areas in the field of architectural design, which reflects regional architectural features to some extent. However, most of the existing methods for architecture form belong to the field of qualitative analysis. Accordingly, quantitative methods are urgently required to extract regional architectural style, identify architecture form, and to and further provide the quantitative evaluation. Based on machine learning technology, this paper proposes a novel method to quantify the feature, form, and evaluation of regional architectures. First, we construct a training dataset—the Chinese Ancient Architecture Image Dataset (CAAID), in which each image is labeled by some experts as having at least one of three typical features such as “High Pedestal”, “Deep Eave” and “Elegant Gable”. Second, the CAAID is used to train our neural network model to identify three kinds of architectural features. In order to reveal the traditional forms of regional architecture in Hubei, we built the Hubei Architectural Heritage Image Dataset (HAHID) as our object dataset, in which we collected architectural images from four different regions including southeast, northeast, southwest, and northwest Hubei. Our object dataset is then fed into our neural network model to predict the typical features for those four regions in Hubei. The obtained quantitative results show that the feature identification of the architectural form is consistent with that of regional architectures in Hubei. Moreover, we can observe from the quantitative results that four geographic regions in Hubei show variation; for instance, the feature of the ‘elegant gable’ in southeastern Hubei is more evident, while the “Deep Eave” in the northwest is more evident. In addition, some new building images are selected to feed into our neural network model and the output quantitative results can effectively identify the corresponding feature style of regional architectures in Hubei. Therefore, our proposed method based on machine learning can be used not only as a quantitative tool to extract features of regional architectures, but also as an effective approach to evaluate architecture forms in the urban renewal process.

APA, Harvard, Vancouver, ISO, and other styles

7

Ma, Rui, Jia-Ching Hsu, Tian Tan, Eriko Nurvitadhi, David Sheffield, Rob Pelt, Martin Langhammer, Jaewoong Sim, Aravind Dasu, and Derek Chiou. "Specializing FGPU for Persistent Deep Learning." ACM Transactions on Reconfigurable Technology and Systems 14, no. 2 (July 8, 2021): 1–23. http://dx.doi.org/10.1145/3457886.

Full text

Abstract:

Overlay architectures are a good way to enable fast development and debug on FPGAs at the expense of potentially limited performance compared to fully customized FPGA designs. When used in concert with hand-tuned FPGA solutions, performant overlay architectures can improve time-to-solution and thus overall productivity of FPGA solutions. This work tunes and specializes FGPU, an open source OpenCL-programmable GPU overlay for FPGAs. We demonstrate that our persistent deep learning (PDL )-FGPU architecture maintains the ease-of-programming and generality of GPU programming while achieving high performance from specialization for the persistent deep learning domain. We also propose an easy method to specialize for other domains. PDL-FGPU includes new instructions, along with micro-architecture and compiler enhancements. We evaluate both the FGPU baseline and the proposed PDL-FGPU on a modern high-end Intel Stratix 10 2800 FPGA in simulation running persistent DL applications (RNN, GRU, LSTM), and non-DL applications to demonstrate generality. PDL-FGPU requires 1.4–3× more ALMs, 4.4–6.4× more M20ks, and 1–9.5× more DSPs than baseline, but improves performance by 56–693× for PDL applications with an average 23.1% degradation on non-PDL applications. We integrated the PDL-FGPU overlay into Intel OPAE to measure real-world performance/power and demonstrate that PDL-FGPU is only 4.0–10.4× slower than the Nvidia V100.

APA, Harvard, Vancouver, ISO, and other styles

8

Sewak, Mohit, Sanjay K. Sahay, and Hemant Rathore. "An Overview of Deep Learning Architecture of Deep Neural Networks and Autoencoders." Journal of Computational and Theoretical Nanoscience 17, no. 1 (January 1, 2020): 182–88. http://dx.doi.org/10.1166/jctn.2020.8648.

Full text

Abstract:

The recent wide applications of deep learning in multiple fields has shown a great progress, but to perform optimally, it requires the adjustment of various architectural features and hyper-parameters. Moreover, deep learning could be used with multiple varieties of architecture aimed at different objectives, e.g., autoencoders are popular for un-supervised learning applications for reducing the dimensionality of the dataset. Similarly, deep neural networks are popular for supervised learning applications viz., classification, regression, etc. Besides the type of deep learning architecture, some other decision criteria and parameter selection decisions are required for determining each layer size, number of layers, activation and loss functions for different layers, optimizer algorithm, regularization, etc. Thus, this paper aims to cover different choices available under each of these major and minor decision criteria for devising a neural network and to train it optimally for achieving the objectives effectively, e.g., malware detection, natural language processing, image recognition, etc.

APA, Harvard, Vancouver, ISO, and other styles

9

Hartanto, Cahyo Adhi, and Laksmita Rahadianti. "Single Image Dehazing Using Deep Learning." JOIV : International Journal on Informatics Visualization 5, no. 1 (March 22, 2021): 76. http://dx.doi.org/10.30630/joiv.5.1.431.

Full text

Abstract:

Many real-world situations such as bad weather may result in hazy environments. Images captured in these hazy conditions will have low image quality due to microparticles in the air. The microparticles light to scatter and absorb, resulting in hazy images with various effects. In recent years, image dehazing has been researched in depth to handle images captured in these conditions. Various methods were developed, from traditional methods to deep learning methods. Traditional methods focus more on the use of statistical prior. These statistical prior have weaknesses in certain conditions. This paper proposes a novel architecture based on PDR-Net by using a pyramid dilated convolution and pre-processing modules, processing modules, post-processing modules, and attention applications. The proposed network is trained to minimize L1 loss and perceptual loss with the O-Haze dataset. To evaluate our architecture's result, we used structural similarity index measure (SSIM), peak signal-to-noise ratio (PSNR), and color difference as an objective assessment and psychovisual experiment as a subjective assessment. Our architecture obtained better results than the previous method using the O-Haze dataset with an SSIM of 0.798, a PSNR of 25.39, but not better on the color difference. The SSIM and PSNR results were strengthened by using subjective assessments and 65 respondents, most of whom chose the results of the restoration of the image produced by our architecture.

APA, Harvard, Vancouver, ISO, and other styles

10

Ghimire, Deepak, Dayoung Kil, and Seong-heum Kim. "A Survey on Efficient Convolutional Neural Networks and Hardware Acceleration." Electronics 11, no. 6 (March 18, 2022): 945. http://dx.doi.org/10.3390/electronics11060945.

Full text

Abstract:

Over the past decade, deep-learning-based representations have demonstrated remarkable performance in academia and industry. The learning capability of convolutional neural networks (CNNs) originates from a combination of various feature extraction layers that fully utilize a large amount of data. However, they often require substantial computation and memory resources while replacing traditional hand-engineered features in existing systems. In this review, to improve the efficiency of deep learning research, we focus on three aspects: quantized/binarized models, optimized architectures, and resource-constrained systems. Recent advances in light-weight deep learning models and network architecture search (NAS) algorithms are reviewed, starting with simplified layers and efficient convolution and including new architectural design and optimization. In addition, several practical applications of efficient CNNs have been investigated using various types of hardware architectures and platforms.

APA, Harvard, Vancouver, ISO, and other styles

11

Zhong, Guoqiang, Kang Zhang, Hongxu Wei, Yuchen Zheng, and Junyu Dong. "Marginal Deep Architecture: Stacking Feature Learning Modules to Build Deep Learning Models." IEEE Access 7 (2019): 30220–33. http://dx.doi.org/10.1109/access.2019.2902631.

Full text

APA, Harvard, Vancouver, ISO, and other styles

12

Sushma, Prof Ksn, Nishant Upadhyay, Ajeet Singh, Prasenjeet Kr Singh, and Tanzeelah Firdaus. "Plant Disease Detection Using Deep Learning." International Journal for Research in Applied Science and Engineering Technology 10, no. 4 (April 30, 2022): 1099–101. http://dx.doi.org/10.22214/ijraset.2022.41451.

Full text

Abstract:

Abstract: Early diagnosis of plant diseases is critical since they have a substantial impact on the growth of their unique species. Many Machine Learning (ML) models have been used to detect and categorize plant diseases, but recent breakthroughs in a subset of ML called Deep Learning (DL) look to hold a lot of promise in terms of improved accuracy. A variety of developed/modified DL architectures, as well as several visualization techniques, are utilized to recognize and identify the symptoms of plant ailments. In addition, a number of performance measurements are used to evaluate various architectures/techniques. This article explains how to use DL models to display a variety of plant diseases. Furthermore, several research gaps are identified, allowing for improved efficiency in detecting plant illnesses even before issues emerge. Keywords: Plant disease; deep learning; convolutional neural networks (CNN), Google Net Architecture, Tensorflow, and PyTorch are some of the tools that can be used;

APA, Harvard, Vancouver, ISO, and other styles

13

Lattari, Francesco, Borja Gonzalez Leon, Francesco Asaro, Alessio Rucci, Claudio Prati, and Matteo Matteucci. "Deep Learning for SAR Image Despeckling." Remote Sensing 11, no. 13 (June 28, 2019): 1532. http://dx.doi.org/10.3390/rs11131532.

Full text

Abstract:

Speckle filtering is an unavoidable step when dealing with applications that involve amplitude or intensity images acquired by coherent systems, such as Synthetic Aperture Radar (SAR). Speckle is a target-dependent phenomenon; thus, its estimation and reduction require the individuation of specific properties of the image features. Speckle filtering is one of the most prominent topics in the SAR image processing research community, who has first tackled this issue using handcrafted feature-based filters. Even if classical algorithms have slowly and progressively achieved better and better performance, the more recent Convolutional-Neural-Networks (CNNs) have proven to be a promising alternative, in the light of the outstanding capabilities in efficiently learning task-specific filters. Currently, only simplistic CNN architectures have been exploited for the speckle filtering task. While these architectures outperform classical algorithms, they still show some weakness in the texture preservation. In this work, a deep encoder–decoder CNN architecture, focused in the specific context of SAR images, is proposed in order to enhance speckle filtering capabilities alongside texture preservation. This objective has been addressed through the adaptation of the U-Net CNN, which has been modified and optimized accordingly. This architecture allows for the extraction of features at different scales, and it is capable of producing detailed reconstructions through its system of skip connections. In this work, a two-phase learning strategy is adopted, by first pre-training the model on a synthetic dataset and by adapting the learned network to the real SAR image domain through a fast fine-tuning procedure. During the fine-tuning phase, a modified version of the total variation (TV) regularization was introduced to improve the network performance when dealing with real SAR data. Finally, experiments were carried out on simulated and real data to compare the performance of the proposed method with respect to the state-of-the-art methodologies.

APA, Harvard, Vancouver, ISO, and other styles

14

Chen, Xihui, Aimin Ji, and Gang Cheng. "A Novel Deep Feature Learning Method Based on the Fused-Stacked AEs for Planetary Gear Fault Diagnosis." Energies 12, no. 23 (November 27, 2019): 4522. http://dx.doi.org/10.3390/en12234522.

Full text

Abstract:

Planetary gear is the key component of the transmission system of electromechanical equipment for energy industry, and it is easy to damage, which affects the reliability and operation efficiency of electromechanical equipment of energy industry. Therefore, it is of great significance to extract the useful fault features and diagnose faults based on raw vibration signals. In this paper, a novel deep feature learning method based on the fused-stacked autoencoders (AEs) for planetary gear fault diagnosis was proposed. First, to improve the data learning ability and the robustness of feature extraction process of AE model, the sparse autoencoder (SAE) and the contractive autoencoder (CAE) were studied, respectively. Then, the quantum ant colony algorithm (QACA) was used to optimize the specific location and key parameters of SAEs and CAEs in deep learning architecture, and multiple SAEs and multiple CAEs were stacked alternately to form a novel deep learning architecture, which gave the deep learning architecture better data learning ability and robustness of feature extraction. The experimental results show that the proposed method can address the raw vibration signals of planetary gear. Compared with other deep learning architectures and shallow learning architecture, the proposed method has better diagnosis performance, and it is an effective method of deep feature learning and fault diagnosis.

APA, Harvard, Vancouver, ISO, and other styles

15

Segura-Bedmar, Isabel, and Pablo Raez. "Cohort selection for clinical trials using deep learning models." Journal of the American Medical Informatics Association 26, no. 11 (September 17, 2019): 1181–88. http://dx.doi.org/10.1093/jamia/ocz139.

Full text

Abstract:

Abstract Objective The goal of the 2018 n2c2 shared task on cohort selection for clinical trials (track 1) is to identify which patients meet the selection criteria for clinical trials. Cohort selection is a particularly demanding task to which natural language processing and deep learning can make a valuable contribution. Our goal is to evaluate several deep learning architectures to deal with this task. Materials and Methods Cohort selection can be formulated as a multilabeling problem whose goal is to determine which criteria are met for each patient record. We explore several deep learning architectures such as a simple convolutional neural network (CNN), a deep CNN, a recurrent neural network (RNN), and CNN-RNN hybrid architecture. Although our architectures are similar to those proposed in existing deep learning systems for text classification, our research also studies the impact of using a fully connected feedforward layer on the performance of these architectures. Results The RNN and hybrid models provide the best results, though without statistical significance. The use of the fully connected feedforward layer improves the results for all the architectures, except for the hybrid architecture. Conclusions Despite the limited size of the dataset, deep learning methods show promising results in learning useful features for the task of cohort selection. Therefore, they can be used as a previous filter for cohort selection for any clinical trial with a minimum of human intervention, thus reducing the cost and time of clinical trials significantly.

APA, Harvard, Vancouver, ISO, and other styles

16

Khrisat, Mohammad S., Anwar Alabadi, Saleh Khawatreh, Majed Omar Al-Dwairi, and Ziad A. Alqadi. "Autoregressive prediction analysis using machine deep learning." Indonesian Journal of Electrical Engineering and Computer Science 27, no. 3 (September 1, 2022): 1509. http://dx.doi.org/10.11591/ijeecs.v27.i3.pp1509-1516.

Full text

Abstract:

Regression analysis, in statistic a modelling, is a set of statical processes that can be used to estimate the relationship between a dependent variable, commonly known as the outcome or response, and more independent variables generally called predictors of covariant. On the other hand, autoregression, which is based on regression equations, is a sequential model that uses time to predict the next step data from the previous step. Given the importance of accurate modelling and reliable predictions. in this paper we have analyzed the most popular methods used for data prediction. Nonlinear autoregressive methods were introduced, and then the machine deep learning approach was used to apply prediction based on a selected input data set. The mean square error was calculated for various artificial neural networks architecture to reach the optimal architecture, which minimized the error. Different artificial neural network (ANN) architectures were trained, tested, and validated using various regressive models, a recommendation was raised according to the obtained and analyzed experimental results. It was shown that using the concepts of machine deep learning will enhance the response of the prediction model.

APA, Harvard, Vancouver, ISO, and other styles

17

Alsaadi, Zaran, Easa Alshamani, Mohammed Alrehaili, Abdulmajeed Ayesh D. Alrashdi, Saleh Albelwi, and Abdelrahman Osman Elfaki. "A Real Time Arabic Sign Language Alphabets (ArSLA) Recognition Model Using Deep Learning Architecture." Computers 11, no. 5 (May 10, 2022): 78. http://dx.doi.org/10.3390/computers11050078.

Full text

Abstract:

Currently, treating sign language issues and producing high quality solutions has attracted researchers and practitioners’ attention due to the considerable prevalence of hearing disabilities around the world. The literature shows that Arabic Sign Language (ArSL) is one of the most popular sign languages due to its rate of use. ArSL is categorized into two groups: The first group is ArSL, where words are represented by signs, i.e., pictures. The second group is ArSl alphabetic (ArSLA), where each Arabic letter is represented by a sign. This paper introduces a real time ArSLA recognition model using deep learning architecture. As a methodology, the proceeding steps were followed. First, a trusted scientific ArSLA dataset was located. Second, the best deep learning architectures were chosen by investigating related works. Third, an experiment was conducted to test the previously selected deep learning architectures. Fourth, the deep learning architecture was selected based on extracted results. Finally, a real time recognition system was developed. The results of the experiment show that the AlexNet architecture is the best due to its high accuracy rate. The model was developed based on AlexNet architecture and successfully tested at real time with a 94.81% accuracy rate.

APA, Harvard, Vancouver, ISO, and other styles

18

Abtahi, Mansour, David Le, Jennifer I. Lim, and Xincheng Yao. "MF-AV-Net: an open-source deep learning network with multimodal fusion options for artery-vein segmentation in OCT angiography." Biomedical Optics Express 13, no. 9 (August 22, 2022): 4870. http://dx.doi.org/10.1364/boe.468483.

Full text

Abstract:

This study is to demonstrate the effect of multimodal fusion on the performance of deep learning artery-vein (AV) segmentation in optical coherence tomography (OCT) and OCT angiography (OCTA); and to explore OCT/OCTA characteristics used in the deep learning AV segmentation. We quantitatively evaluated multimodal architectures with early and late OCT-OCTA fusions, compared to the unimodal architectures with OCT-only and OCTA-only inputs. The OCTA-only architecture, early OCT-OCTA fusion architecture, and late OCT-OCTA fusion architecture yielded competitive performances. For the 6 mm×6 mm and 3 mm×3 mm datasets, the late fusion architecture achieved an overall accuracy of 96.02% and 94.00%, slightly better than the OCTA-only architecture which achieved an overall accuracy of 95.76% and 93.79%. 6 mm×6 mm OCTA images show AV information at pre-capillary level structure, while 3 mm×3 mm OCTA images reveal AV information at capillary level detail. In order to interpret the deep learning performance, saliency maps were produced to identify OCT/OCTA image characteristics for AV segmentation. Comparative OCT and OCTA saliency maps support the capillary-free zone as one of the possible features for AV segmentation in OCTA. The deep learning network MF-AV-Net used in this study is available on GitHub for open access.

APA, Harvard, Vancouver, ISO, and other styles

19

Pachón, César G., Dora M. Ballesteros, and Diego Renza. "Fake Banknote Recognition Using Deep Learning." Applied Sciences 11, no. 3 (January 30, 2021): 1281. http://dx.doi.org/10.3390/app11031281.

Full text

Abstract:

Recently, some state-of-the-art works have used deep learning-based architectures, specifically convolutional neural networks (CNNs), for banknote recognition and counterfeit detection with promising results. However, it is not clear which design strategy is more appropriate (custom or by transfer learning) in terms of classifier performance and inference times for massive data applications. This paper presents a comparison of the two design strategies in various types of architecture. For the transfer learning (TL) strategy, the most appropriate freezing points in CNN architectures (sequential, residual and Inception) are identified. In addition, a custom model based on an AlexNet-type sequential CNN is proposed. Both the TL and the custom models were trained and compared using a Colombian banknote dataset. According to the results, ResNet18 achieved the best accuracy, with 100%. On the other hand, the network with the shortest inference times was the proposed custom network, since its performance is up to 6.48-times faster in CPU and 16.29-times faster in GPU than the inference time with the models by transfer learning.

APA, Harvard, Vancouver, ISO, and other styles

20

Cheng, Anda, Jiaxing Wang, Xi Sheryl Zhang, Qiang Chen, Peisong Wang, and Jian Cheng. "DPNAS: Neural Architecture Search for Deep Learning with Differential Privacy." Proceedings of the AAAI Conference on Artificial Intelligence 36, no. 6 (June 28, 2022): 6358–66. http://dx.doi.org/10.1609/aaai.v36i6.20586.

Full text

Abstract:

Training deep neural networks (DNNs) for meaningful differential privacy (DP) guarantees severely degrades model utility. In this paper, we demonstrate that the architecture of DNNs has a significant impact on model utility in the context of private deep learning, whereas its effect is largely unexplored in previous studies. In light of this missing, we propose the very first framework that employs neural architecture search to automatic model design for private deep learning, dubbed as DPNAS. To integrate private learning with architecture search, a DP-aware approach is introduced for training candidate models composed on a delicately defined novel search space. We empirically certify the effectiveness of the proposed framework. The searched model DPNASNet achieves state-of-the-art privacy/utility trade-offs, e.g., for the privacy budget of (epsilon, delta)=(3, 1e-5), our model obtains test accuracy of 98.57% on MNIST, 88.09% on FashionMNIST, and 68.33% on CIFAR-10. Furthermore, by studying the generated architectures, we provide several intriguing findings of designing private-learning-friendly DNNs, which can shed new light on model design for deep learning with differential privacy.

APA, Harvard, Vancouver, ISO, and other styles

21

Pollok, Stefan, and Rasmus Bjørk. "Deep learning for magnetism." Europhysics News 53, no. 2 (2022): 18–21. http://dx.doi.org/10.1051/epn/2022204.

Full text

Abstract:

In deep learning, neural networks consisting of trainable parameters are designed to model unknown functions based on available data. When the underlying physics of the system at hand are known, e.g., Maxwell’s equation in electromagnetism, then these can be embedded into the deep learning architecture to obtain better function approximations.

APA, Harvard, Vancouver, ISO, and other styles

22

Belhaouari, Samir Brahim, and Hafsa Raissouli. "MADL: A Multilevel Architecture of Deep Learning." International Journal of Computational Intelligence Systems 14, no. 1 (2021): 693. http://dx.doi.org/10.2991/ijcis.d.201216.003.

Full text

APA, Harvard, Vancouver, ISO, and other styles

23

Lee, Soo-Hwan, Jong-Chan Kim, and Dong-Hoan Seo. "Image reconstruction technique using deep learning architecture." Journal of the Korean Society of Marine Engineering 42, no. 2 (February 28, 2018): 121–26. http://dx.doi.org/10.5916/jkosme.2018.42.2.121.

Full text

APA, Harvard, Vancouver, ISO, and other styles

24

Ayad, Hayder, Ikhlas Watan Ghindawi, and Mustafa Salam Kadhm. "Lung Segmentation Using Proposed Deep Learning Architecture." International Journal of Online and Biomedical Engineering (iJOE) 16, no. 15 (December 15, 2020): 141. http://dx.doi.org/10.3991/ijoe.v16i15.17115.

Full text

Abstract:

<div id="titleAndAbstract"><table class="data" width="100%"><tbody><tr valign="top"><td class="value">The Prediction and detection disease in human lungs are a very critical operation. It depends on an efficient view of the CT images to the doctors. It depends on an efficient view of the CT images to the doctors. The clear view of the images to clearly identify the disease depends on the segmentation that may save people lives. Therefore, an accurate lung segmentation system from CT image based on proposed CNN architecture is proposed. The system used weighted softmax function the improved the segmentation accuracy. By experiments, the system achieved a high segmentation accuracy 98.9% using LIDC-IDRI CT lung images database.</td></tr></tbody></table></div><div id="indexing"> </div>

APA, Harvard, Vancouver, ISO, and other styles

25

Chen, Xi-liang, Lei Cao, Chen-xi Li, Zhi-xiong Xu, and Jun Lai. "Ensemble Network Architecture for Deep Reinforcement Learning." Mathematical Problems in Engineering 2018 (2018): 1–6. http://dx.doi.org/10.1155/2018/2129393.

Full text

Abstract:

The popular deepQlearning algorithm is known to be instability because of theQ-value’s shake and overestimation action values under certain conditions. These issues tend to adversely affect their performance. In this paper, we develop the ensemble network architecture for deep reinforcement learning which is based on value function approximation. The temporal ensemble stabilizes the training process by reducing the variance of target approximation error and the ensemble of target values reduces the overestimate and makes better performance by estimating more accurateQ-value. Our results show that this architecture leads to statistically significant better value evaluation and more stable and better performance on several classical control tasks at OpenAI Gym environment.

APA, Harvard, Vancouver, ISO, and other styles

26

Verma, Arnav, Devesh Samaiya, and Karunesh K. Gupta. "Nonlinear Motion Tracking by Deep Learning Architecture." IOP Conference Series: Materials Science and Engineering 331 (March 2018): 012020. http://dx.doi.org/10.1088/1757-899x/331/1/012020.

Full text

APA, Harvard, Vancouver, ISO, and other styles

27

Pon Kumar, Steven Spielberg, Aditya Tulsyan, Bhushan Gopaluni, and Philip Loewen. "A Deep Learning Architecture for Predictive Control." IFAC-PapersOnLine 51, no. 18 (2018): 512–17. http://dx.doi.org/10.1016/j.ifacol.2018.09.373.

Full text

APA, Harvard, Vancouver, ISO, and other styles

28

Li, Xiang, Ling Peng, Yuan Hu, Jing Shao, and Tianhe Chi. "Deep learning architecture for air quality predictions." Environmental Science and Pollution Research 23, no. 22 (October 13, 2016): 22408–17. http://dx.doi.org/10.1007/s11356-016-7812-9.

Full text

APA, Harvard, Vancouver, ISO, and other styles

29

Siłka, Wojciech, Michał Wieczorek, Jakub Siłka, and Marcin Woźniak. "Malaria Detection Using Advanced Deep Learning Architecture." Sensors 23, no. 3 (January 29, 2023): 1501. http://dx.doi.org/10.3390/s23031501.

Full text

Abstract:

Malaria is a life-threatening disease caused by parasites that are transmitted to humans through the bites of infected mosquitoes. The early diagnosis and treatment of malaria are crucial for reducing morbidity and mortality rates, particularly in developing countries where the disease is prevalent. In this article, we present a novel convolutional neural network (CNN) architecture for detecting malaria from blood samples with a 99.68% accuracy. Our method outperforms the existing approaches in terms of both accuracy and speed, making it a promising tool for malaria diagnosis in resource-limited settings. The CNN was trained on a large dataset of blood smears and was able to accurately classify infected and uninfected samples with high sensitivity and specificity. Additionally, we present an analysis of model performance on different subtypes of malaria and discuss the implications of our findings for the use of deep learning in infectious disease diagnosis.

APA, Harvard, Vancouver, ISO, and other styles

30

P, Shanmugavadivu, Mary Shanthi Rani M, Chitra P, Lakshmanan S, Nagaraja P, and Vignesh U. "Bio-Optimization of Deep Learning Network Architectures." Security and Communication Networks 2022 (September 20, 2022): 1–11. http://dx.doi.org/10.1155/2022/3718340.

Full text

Abstract:

Deep learning is reaching new heights as a result of its cutting-edge performance in a variety of fields, including computer vision, natural language processing, time series analysis, and healthcare. Deep learning is implemented using batch and stochastic gradient descent methods, as well as a few optimizers; however, this led to subpar model performance. However, there is now a lot of effort being done to improve deep learning’s performance using gradient optimization methods. The suggested work analyses convolutional neural networks (CNN) and deep neural networks (DNN) using several cutting-edge optimizers to enhance the performance of architectures. This work uses specific optimizers (SGD, RMSprop, Adam, Adadelta, etc.) to enhance the performance of designs using different types of datasets for result matching. A thorough report on the optimizers’ performance across a variety of architectures and datasets finishes the study effort. This research will be helpful to researchers in developing their framework and appropriate architecture optimizers. The proposed work involves eight new optimizers using four CNN and DNN architectures. The experimental results exploit breakthrough results for improving the efficiency of CNN and DNN architectures using various datasets.

APA, Harvard, Vancouver, ISO, and other styles

31

Karypidis, Efstathios, Stylianos G. Mouslech, Kassiani Skoulariki, and Alexandros Gazis. "Comparison Analysis of Traditional Machine Learning and Deep Learning Techniques for Data and Image Classification." WSEAS TRANSACTIONS ON MATHEMATICS 21 (March 23, 2022): 122–30. http://dx.doi.org/10.37394/23206.2022.21.19.

Full text

Abstract:

The purpose of the study is to analyse and compare the most common machine learning and deep learning techniques used for computer vision 2D object classification tasks. Firstly, we will present the theoretical background of the Bag of Visual words model and Deep Convolutional Neural Networks (DCNN). Secondly, we will implement a Bag of Visual Words model, the VGG16 CNN Architecture. Thirdly, we will present our custom and novice DCNN in which we test the aforementioned implementations on a modified version of the Belgium Traffic Sign dataset. Our results showcase the effects of hyperparameters on traditional machine learning and the advantage in terms of accuracy of DCNNs compared to classical machine learning methods. As our tests indicate, our proposed solution can achieve similar - and in some cases better - results than existing DCNNs architectures. Finally, the technical merit of this article lies in the presented computationally simpler DCNN architecture, which we believe can pave the way towards using more efficient architectures for basic tasks.

APA, Harvard, Vancouver, ISO, and other styles

32

Velankar, Makarand, Sneha Thombre, and Harshad Wadkar. "EVALUATING DEEP LEARNING MODELS FOR MUSIC EMOTION RECOGNITION." International Journal of Engineering Applied Sciences and Technology 7, no. 6 (October 1, 2022): 252–59. http://dx.doi.org/10.33564/ijeast.2022.v07i06.026.

Full text

Abstract:

Music listening helps people not only for entertainment, but also to reduce emotional stress in their daily lives. People nowadays tend to use online music streaming services such as Spotify, Amazon Music, Google Play Music, etc. rather than storing the songs on their devices. The songs in these streaming services are categorized into different emotional labels such as happy, sad, romantic, devotional, etc. In the music streaming applications, the songs are manually tagged with their emotional categories for music recommendation. Considering the growth of music on different social media platforms and the internet, the need for automatic tagging will increase in coming time. The work presented deals with training the deep learning model for automatic emotional tagging. It covers implementation of two different deep learning architectures for classifying the audio files using the Mel-spectrogram of music audio. The first architecture proposed is Convolutional Recurrent Model (CRNN) and the second architecture is a Parallel Convolutional Recurrent Model (Parallel CNN). Both the architectures exploit the combined features of Convolutional and Recurrent layers. This combination is used to extract features from time and frequency domains. The results with accuracies in the range of 51 to 54 % are promising for both models for a small dataset of 138 songs, considering the large datasets required for training deep learning models.

APA, Harvard, Vancouver, ISO, and other styles

33

Billah, Umme Hafsa, Hung Manh La, and Alireza Tavakkoli. "Deep Learning-Based Feature Silencing for Accurate Concrete Crack Detection." Sensors 20, no. 16 (August 7, 2020): 4403. http://dx.doi.org/10.3390/s20164403.

Full text

Abstract:

An autonomous concrete crack inspection system is necessary for preventing hazardous incidents arising from deteriorated concrete surfaces. In this paper, we present a concrete crack detection framework to aid the process of automated inspection. The proposed approach employs a deep convolutional neural network architecture for crack segmentation, while addressing the effect of gradient vanishing problem. A feature silencing module is incorporated in the proposed framework, capable of eliminating non-discriminative feature maps from the network to improve performance. Experimental results support the benefit of incorporating feature silencing within a convolutional neural network architecture for improving the network’s robustness, sensitivity, and specificity. An added benefit of the proposed architecture is its ability to accommodate for the trade-off between specificity (positive class detection accuracy) and sensitivity (negative class detection accuracy) with respect to the target application. Furthermore, the proposed framework achieves a high precision rate and processing time than the state-of-the-art crack detection architectures.

APA, Harvard, Vancouver, ISO, and other styles

34

Javaid, Sameena, Safdar Rizvi, Muhammad Talha Ubaid, Abdou Darboe, and Shakir Mahmood Mayo. "Interpretation of Expressions through Hand Signs Using Deep Learning Techniques." Vol 4 Issue 2 4, no. 2 (June 25, 2022): 596–611. http://dx.doi.org/10.33411/ijist/2022040225.

Full text

Abstract:

It is a challenging task to interpret sign language automatically, as it comprises high-level vision features to accurately understand and interpret the meaning of the signer or vice versa. In the current study, we automatically distinguish hand signs and classify seven basic gestures representing symbolic emotions or expressions like happy, sad, neutral, disgust, scared, anger, and surprise. Convolutional Neural Network is a famous method for classifications using vision-based deep learning; here in the current study, proposed transfer learning using a well-known architecture of VGG16 to speed up the convergence and improve accuracy by using pre-trained weights. We obtained a high accuracy of 99.98% of the proposed architecture with a minimal and low-quality data set of 455 images collected by 65 individuals for seven hand gesture classes. Further, compared the performance of VGG16 architecture with two different optimizers, SGD, and Adam, along with some more architectures of Alex Net, LeNet05, and ResNet50.

APA, Harvard, Vancouver, ISO, and other styles

35

Akbani, Sufiyan Salim, Adeeba Naaz, Nazish Kausar, and Prof Abdul Razzaque. "Brain Tumor Detection Using Deep Learning." International Journal for Research in Applied Science and Engineering Technology 10, no. 4 (April 30, 2022): 573–77. http://dx.doi.org/10.22214/ijraset.2022.41321.

Full text

Abstract:

Abstract: One of the most leading death causes in the world is brain tumor. Tumor Detection is one of the most difficult tasks in medical image processing. In fact, the manual classification with human-assisted support can be improper prediction and diagnosis shown by medical evidence. The detection task is too difficult to perform because there is a lot of diversity in the images as brain tumors come in different shapes and textures. Recently, deep learning techniques showed promising results towards improving accuracy of detection and classification of brain tumor from magnetic resonance imaging (MRI). In this paper, we propose a deep learning model for the classification of brain tumors from MRI images using convolutional neural network (CNN) based on transfer learning. The implemented system explores a number of CNN architectures, image preprocessing and transfer learning model named MobilNet to achieve the better performance and accuracy. Keywords: Deep learning, convolutional neural network, Transfer learning, Brain tumor, medical image classification, MobileNet architecture, etc.

APA, Harvard, Vancouver, ISO, and other styles

36

Farhi, Lubna, Saadia Mansoor Kazmi, Hassan Imam, Mejdal Alqahtani, and Farhan Ur Rehman. "Dermoscopic Image Classification Using Deep Belief Learning Network Architecture." Wireless Communications and Mobile Computing 2022 (May 25, 2022): 1–13. http://dx.doi.org/10.1155/2022/2415726.

Full text

Abstract:

In this paper, deep belief learning network architecture (DBL) is proposed for medical image classification in a bid to improve the diagnostics of dermal melanoma as an alternative to traditional dermoscopy. Preprocessing was carried out by using a linear Gaussian filter by eliminating high-frequency artifacts and distortion. The K -means segmentation technique was used to extract the region of interest. The DBL network was then applied to the segmented image for classification. The DBL architecture disperses the weights and hyperparameters to all positions in an image, making it possible to scale to various image sizes. The effects of overfitting were mitigated for small datasets and were achieved by optimizing the proposed network. The algorithm works effectively by fine-tuning constraints. The results showed an increase in the accuracy between the proposed model and AlexNet and LeeNet for segmented images from 8% to 47%, respectively. Similarly, an increase for nonsegmented images was observed between 2% and 48%. An average reduction of 47.8% and 41.5% in error for both segmented and nonsegmented images was recorded for dermal images. The execution time also decreased in comparison with the other architectures averaged by 8-13%, since the weights were distributed only on the clustered regions in the segmented image, as compared to the whole image thus allowing the network to classify it faster with improved accuracy.

APA, Harvard, Vancouver, ISO, and other styles

37

Defresne, Marianne, Sophie Barbe, and Thomas Schiex. "Protein Design with Deep Learning." International Journal of Molecular Sciences 22, no. 21 (October 29, 2021): 11741. http://dx.doi.org/10.3390/ijms222111741.

Full text

Abstract:

Computational Protein Design (CPD) has produced impressive results for engineering new proteins, resulting in a wide variety of applications. In the past few years, various efforts have aimed at replacing or improving existing design methods using Deep Learning technology to leverage the amount of publicly available protein data. Deep Learning (DL) is a very powerful tool to extract patterns from raw data, provided that data are formatted as mathematical objects and the architecture processing them is well suited to the targeted problem. In the case of protein data, specific representations are needed for both the amino acid sequence and the protein structure in order to capture respectively 1D and 3D information. As no consensus has been reached about the most suitable representations, this review describes the representations used so far, discusses their strengths and weaknesses, and details their associated DL architecture for design and related tasks.

APA, Harvard, Vancouver, ISO, and other styles

38

Fielding, Ben, and Li Zhang. "Evolving Deep DenseBlock Architecture Ensembles for Image Classification." Electronics 9, no. 11 (November 9, 2020): 1880. http://dx.doi.org/10.3390/electronics9111880.

Full text

Abstract:

Automatic deep architecture generation is a challenging task, owing to the large number of controlling parameters inherent in the construction of deep networks. The combination of these parameters leads to the creation of large, complex search spaces that are feasibly impossible to properly navigate without a huge amount of resources for parallelisation. To deal with such challenges, in this research we propose a Swarm Optimised DenseBlock Architecture Ensemble (SODBAE) method, a joint optimisation and training process that explores a constrained search space over a skeleton DenseBlock Convolutional Neural Network (CNN) architecture. Specifically, we employ novel weight inheritance learning mechanisms, a DenseBlock skeleton architecture, as well as adaptive Particle Swarm Optimisation (PSO) with cosine search coefficients to devise networks whilst maintaining practical computational costs. Moreover, the architecture design takes advantage of recent advancements of the concepts of residual connections and dense connectivity, in order to yield CNN models with a much wider variety of structural variations. The proposed weight inheritance learning schemes perform joint optimisation and training of the architectures to reduce the computational costs. Being evaluated using the CIFAR-10 dataset, the proposed model shows great superiority in classification performance over other state-of-the-art methods while illustrating a greater versatility in architecture generation.

APA, Harvard, Vancouver, ISO, and other styles

39

Sen, Gabriel, Albert Adeboye, and Oluwole Alagbe. "The Influence of Architecture Students’ Learning Approaches on their Academic Performance in Two Nigeria Universities." International Journal of Learning, Teaching and Educational Research 20, no. 2 (February 28, 2021): 137–51. http://dx.doi.org/10.26803/ijlter.20.2.8.

Full text

Abstract:

The paper was a pilot study that examined learning approaches of architecture students; variability of approaches by university type and gender and; influence of architecture students’ learning approaches on their academic performance. The sample was 349 architecture students from two universities. Descriptive and statistical analyses were used. Results revealed predominant use of deep learning approaches by students. Furthermore, learning approaches neither significantly differed by university type nor gender. Regression analysis revealed that demographic factors accounted for 2.9% of variation in academic performance (F (2,346) = 6.2, p = 0.002, R2 = 0.029, f2 = 0.029) and when learning approaches were also entered the model accounted for 4.4% of variation in academic performance (F (14,334) =2.2, p =0.009, R2 = 0.044, f2=0.044). Deep learning approaches significantly and positively influenced variation in academic performance while surface learning approaches significantly and negatively influenced academic performance. This implies that architectural educators should use instructional methods that encourage deep approaches. Future research needs to use larger and more heterogeneous samples for confirmation of results.

APA, Harvard, Vancouver, ISO, and other styles

40

Ruder, Sebastian, Joachim Bingel, Isabelle Augenstein, and Anders Søgaard. "Latent Multi-Task Architecture Learning." Proceedings of the AAAI Conference on Artificial Intelligence 33 (July 17, 2019): 4822–29. http://dx.doi.org/10.1609/aaai.v33i01.33014822.

Full text

Abstract:

Multi-task learning (MTL) allows deep neural networks to learn from related tasks by sharing parameters with other networks. In practice, however, MTL involves searching an enormous space of possible parameter sharing architectures to find (a) the layers or subspaces that benefit from sharing, (b) the appropriate amount of sharing, and (c) the appropriate relative weights of the different task losses. Recent work has addressed each of the above problems in isolation. In this work we present an approach that learns a latent multi-task architecture that jointly addresses (a)–(c). We present experiments on synthetic data and data from OntoNotes 5.0, including four different tasks and seven different domains. Our extension consistently outperforms previous approaches to learning latent architectures for multi-task problems and achieves up to 15% average error reductions over common approaches to MTL.

APA, Harvard, Vancouver, ISO, and other styles

41

Hernandez-Leal, Pablo, Bilal Kartal, and Matthew E. Taylor. "Agent Modeling as Auxiliary Task for Deep Reinforcement Learning." Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment 15, no. 1 (October 8, 2019): 31–37. http://dx.doi.org/10.1609/aiide.v15i1.5221.

Full text

Abstract:

In this paper we explore how actor-critic methods in deep reinforcement learning, in particular Asynchronous Advantage Actor-Critic (A3C), can be extended with agent modeling. Inspired by recent works on representation learning and multiagent deep reinforcement learning, we propose two architectures to perform agent modeling: the first one based on parameter sharing, and the second one based on agent policy features. Both architectures aim to learn other agents’ policies as auxiliary tasks, besides the standard actor (policy) and critic (values). We performed experiments in both cooperative and competitive domains. The former is a problem of coordinated multiagent object transportation and the latter is a two-player mini version of the Pommerman game. Our results show that the proposed architectures stabilize learning and outperform the standard A3C architecture when learning a best response in terms of expected rewards.

APA, Harvard, Vancouver, ISO, and other styles

42

Ahamad, Afaroj, Chi-Chia Sun, and Wen-Kai Kuo. "Quantized Semantic Segmentation Deep Architecture for Deployment on an Edge Computing Device for Image Segmentation." Electronics 11, no. 21 (October 31, 2022): 3561. http://dx.doi.org/10.3390/electronics11213561.

Full text

Abstract:

In the field of computer vision technology, deep learning of image processing has become an emerging research area. The semantic segmentation of an image is among the utmost essential and significant tasks in image-processing research, offering a wide range of application fields such as autonomous driving systems, medical diagnosis, surveillance security, etc. Thus far, many studies have suggested and developed neural network modules in deep learning. To the best of our knowledge, all existing neural networks for semantic segmentation have large parameter sizes and it is therefore unfeasible to implement those architectures in low-power and memory-limited embedded platforms such as FPGAs. Building an embedded platform with that architecture is possible after reducing the parameter size without affecting the module’s architecture. The quantization technique lowers the precision of the neural network parameters while mostly keeping the accuracy. In this paper, we propose a quantization algorithm for a semantic segmentation deep learning architecture, which reduces the parameter size by four to eight times with a negligible accuracy abatement. As long as the parameter size is reduced, the deep learning architecture is improved in terms of required storage, computational speed, and power efficiency.

APA, Harvard, Vancouver, ISO, and other styles

43

Nguyen, Chi Cuong, Giang Son Tran, Thi Phuong Nghiem, Jean-Christophe Burie, and Chi Mai Luong. "Real-Time Smile Detection using Deep Learning." Journal of Computer Science and Cybernetics 35, no. 2 (June 3, 2019): 135–45. http://dx.doi.org/10.15625/1813-9663/35/2/13315.

Full text

Abstract:

Real-time smile detection from facial images is useful in many real world applications such as automatic photo capturing in mobile phone cameras or interactive distance learning. In this paper, we study different architectures of object detection deep networks for solving real-time smile detection problem. We then propose a combination of a lightweight convolutional neural network architecture (BKNet) with an efficient object detection framework (RetinaNet). The evaluation on the two datasets (GENKI-4K, UCF Selfie) with a mid-range hardware device (GTX TITAN Black) show that our proposed method helps in improving both accuracy and inference time of the original RetinaNet to reach real-time performance. In comparison with the state-of-the-art object detection framework (YOLO), our method has higher inference time, but still reaches real-time performance and obtains higher accuracy of smile detection on both experimented datasets.

APA, Harvard, Vancouver, ISO, and other styles

44

Liu, Xiaobo, Chaochao Zhang, Zhihua Cai, Jianfeng Yang, Zhilang Zhou, and Xin Gong. "Continuous Particle Swarm Optimization-Based Deep Learning Architecture Search for Hyperspectral Image Classification." Remote Sensing 13, no. 6 (March 12, 2021): 1082. http://dx.doi.org/10.3390/rs13061082.

Full text

Abstract:

Deep convolutional neural networks (CNNs) are widely used in hyperspectral image (HSI) classification. However, the most successful CNN architectures are handcrafted, which need professional knowledge and consume a very significant amount of time. To automatically design cell-based CNN architectures for HSI classification, we propose an efficient continuous evolutionary method, named CPSO-Net, which can dramatically accelerate optimal architecture generation by the optimization of weight-sharing parameters. First, a SuperNet with all candidate operations is maintained to share the parameters for all individuals and optimized by collecting the gradients of all individuals in the population. Second, a novel direct encoding strategy is devised to encode architectures into particles, which inherit the parameters from the SuperNet. Then, particle swarm optimization is used to search for the optimal deep architecture from the particle swarm. Furthermore, experiments with limited training samples based on four widely used biased and unbiased hyperspectral datasets showed that our proposed method achieves good performance comparable to the state-of-the-art HSI classification methods.

APA, Harvard, Vancouver, ISO, and other styles

45

Nistor, Sergiu Cosmin, Tudor Alexandru Ileni, and Adrian Sergiu Dărăbant. "Automatic Development of Deep Learning Architectures for Image Segmentation." Sustainability 12, no. 22 (November 20, 2020): 9707. http://dx.doi.org/10.3390/su12229707.

Full text

Abstract:

Machine learning is a branch of artificial intelligence that has gained a lot of traction in the last years due to advances in deep neural networks. These algorithms can be used to process large quantities of data, which would be impossible to handle manually. Often, the algorithms and methods needed for solving these tasks are problem dependent. We propose an automatic method for creating new convolutional neural network architectures which are specifically designed to solve a given problem. We describe our method in detail and we explain its reduced carbon footprint, computation time and cost compared to a manual approach. Our method uses a rewarding mechanism for creating networks with good performance and so gradually improves its architecture proposals. The application for the algorithm that we chose for this paper is segmentation of eyeglasses from images, but our method is applicable, to a larger or lesser extent, to any image processing task. We present and discuss our results, including the architecture that obtained 0.9683 intersection-over-union (IOU) score on our most complex dataset.

APA, Harvard, Vancouver, ISO, and other styles

46

Rathnam, S. Muni, and G. Siva Koteswara Rao. "A Novel Deep Learning Architecture for Image Hiding." WSEAS TRANSACTIONS ON SIGNAL PROCESSING 16 (February 26, 2021): 206–10. http://dx.doi.org/10.37394/232014.2020.16.23.

Full text

Abstract:

Watermarking is a today's digital hiding technique within certain electronic content: for example, message, image, video, or audio recordings. Recent times, it was created as a modern copyright security tool. The pattern in zero watermarking technique isn't really inserted directly in the cover image, but has a logical relation with that cover image. In this article, we propose a powerful convolution neural Networks (CNN) and deep learning algorithm-based-watermarking technique in which the CNN produces robust inherent selected features and is merged with the XOR activity of host's watermark sequence. The outcomes of our proposed method present the courage of the watermark counter to many typical image processing techniques.

APA, Harvard, Vancouver, ISO, and other styles

47

Bobadilla, Jesus, Santiago Alonso, and Antonio Hernando. "Deep Learning Architecture for Collaborative Filtering Recommender Systems." Applied Sciences 10, no. 7 (April 3, 2020): 2441. http://dx.doi.org/10.3390/app10072441.

Full text

Abstract:

This paper provides an innovative deep learning architecture to improve collaborative filtering results in recommender systems. It exploits the potential of the reliability concept to raise predictions and recommendations quality by incorporating prediction errors (reliabilities) in the deep learning layers. The underlying idea is to recommend highly predicted items that also have been found as reliable ones. We use the deep learning architecture to extract the existing non-linear relations between predictions, reliabilities, and accurate recommendations. The proposed architecture consists of three related stages, providing three stacked abstraction levels: (a) real prediction errors, (b) predicted errors (reliabilities), and (c) predicted ratings (predictions). In turn, each abstraction level requires a learning process: (a) Matrix Factorization from ratings, (b) Multilayer Neural Network fed with real prediction errors and hidden factors, and (c) Multilayer Neural Network fed with reliabilities and hidden factors. A complete set of experiments has been run involving three representative and open datasets and a state-of-the-art baseline. The results show strong prediction improvements and also important recommendation improvements, particularly for the recall quality measure.

APA, Harvard, Vancouver, ISO, and other styles

48

Sun, Maoran, Fan Zhang, Fabio Duarte, and Carlo Ratti. "Understanding architecture age and style through deep learning." Cities 128 (September 2022): 103787. http://dx.doi.org/10.1016/j.cities.2022.103787.

Full text

APA, Harvard, Vancouver, ISO, and other styles

49

Chakravarthy, Arnav. "HYBRID ARCHITECTURE FOR SENTIMENT ANALYSIS USING DEEP LEARNING." International Journal of Advanced Research in Computer Science 9, no. 1 (February 20, 2018): 735–38. http://dx.doi.org/10.26483/ijarcs.v9i1.5388.

Full text

APA, Harvard, Vancouver, ISO, and other styles

50

Bui, Trong-An, Pei-Jun Lee, Kai-Yew Lum, Clarissa Loh, and Kyo Tan. "Deep Learning for Landslide Recognition in Satellite Architecture." IEEE Access 8 (2020): 143665–78. http://dx.doi.org/10.1109/access.2020.3014305.

Full text

APA, Harvard, Vancouver, ISO, and other styles

Journal articles on the topic 'Deep learning architecture'

Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles